1. Field of the Invention
The present invention relates to a device and a method for selectively blending two images processed with two different methods.
2. Description of the Related Art
Error diffusion processing, screen processing and the like have been proposed as processing for performing halftone conversion (dithering) on an image.
Reproducing images with error diffusion processing results in greater sharpness than when images are reproduced with screen processing, although graininess increases.
In view of this, Japanese Laid-open Patent Publication No. 2003-274173 proposes a method for making images reproduced with error diffusion processing smoother than has been possible to date.
Also, Japanese Laid-open Patent Publication No. 2004-179768 proposes a method for selectively using error diffusion processing and screen processing according to an image processing mode. According to the method disclosed in Japanese Laid-open Patent Publication No. 2004-179768, output resulting from error diffusion processing is selected as a default if the image processing mode is “text/photo blend mode” or “text mode”. Output resulting from screen processing of 141 lpi halftone dots is selected as a default if in “photo mode” (para 0012 to 0014).
FIG. 10 shows an example of an unnatural connection between dots.
However, it may be desired to divide a single image into a plurality of regions, and reproduce some of the regions by error diffusion processing and reproduce the remaining regions by screen processing. In such a case, an unnatural connection between dots such as shown in FIG. 10 may appear near the boundary between an error-diffused region and a screen-processed region. Such an unnatural connection between dots produces image unevenness at the boundary portion, and is a cause of image degradation.